Supermarket sales data: a tool for measuring regional differences in dietary habits.

OBJECTIVE To assess how well supermarket sales data from a major supermarket chain can reflect on regional differences in dietary behaviour by comparing the sales data with the results provided by the annual health behaviour surveys. DESIGN Cross-sectional observational study. SETTING The study was carried out in six Finnish cities situated in different parts of Finland. For the study supermarket sales data of milk, sour milk, fats and oils for 1 month, September 1997, were obtained from eight supermarkets. Proportional sales of different types of dairy products were calculated as well as mean salt and fat per cent and the proportion of saturated fat to total fat. The health behaviour surveys from spring 1995, 1996 and 1997 provided information about dietary habits of the adult population in the cities. The reported use of dairy products was compared with the percentage sales. RESULTS The proportional sales of dairy products varied between the cities. In Pori in western Finland the sale of milk fat was highest in all food groups. In Oulu, northern Finland, the sale of non-fat milk was high. In the capital region the sale of oil was highest. Regional differences could also be seen in the survey data. The similarity between the two different datasets was high. CONCLUSIONS The use of supermarket sales data for assessing regional differences in health behaviour is feasible. The challenge will be to get supermarket managers willing to provide sales data on a routine basis for monitoring and research.